I'm working with a Python regex for extracting time durations in '2h30m' format. I run into an issue where non-capturing groups ((?:...)) are getting captured inside named groups.
我正在使用Python正则表达式以'2h30m'格式提取持续时间。我遇到了一个问题，其中非捕获组（（？：...））被捕获在命名组内。
e.g. matching 2h30m against:
例如匹配2h30m对：
(?P<hours>\d+(?:h))?(?P<minutes>\d+(?:m))?
would match {'hours': '2h', 'minutes': '30m'}, and not 2 and 30.
将匹配{'小时'：'2h'，'分钟'：'30m'}，而不是2和30。
The workaround would be to use a positive lookahead assertions ((?=...)), but this doesn't update the state of the regex FSM so we have to repeat the h, m suffixes:
解决方法是使用正向前瞻断言（（？= ...）），但这不会更新正则表达式FSM的状态，因此我们必须重复h，m后缀：
(?P<hours>\d+(?=h))?h?(?P<minutes>\d+(?=m))?m?
Is there a better way to do this?
有一个更好的方法吗？
1 个解决方案
#1
7
Non-capturing groups don't "anti-capture" what they match and remove them from outer groups. They're just a way to group things together so you can apply quantifiers to them.
非捕获组不会“反捕获”它们匹配的内容并将其从外部组中删除。它们只是将事物分组在一起的一种方式，因此您可以将量词应用于它们。
To get the effect you want, you can rearrange the groups to put the non-capturing groups outside the capturing groups:
要获得所需的效果，可以重新排列组以将非捕获组置于捕获组之外：
(?:(?P<hours>\d+)h)?(?:(?P<minutes>\d+)m)?
#1
7
Non-capturing groups don't "anti-capture" what they match and remove them from outer groups. They're just a way to group things together so you can apply quantifiers to them.
非捕获组不会“反捕获”它们匹配的内容并将其从外部组中删除。它们只是将事物分组在一起的一种方式，因此您可以将量词应用于它们。
To get the effect you want, you can rearrange the groups to put the non-capturing groups outside the capturing groups:
要获得所需的效果，可以重新排列组以将非捕获组置于捕获组之外：
(?:(?P<hours>\d+)h)?(?:(?P<minutes>\d+)m)?